Hey guys, first post on the forum, I did a brief look-over and I don't think anyone is discussing this, so I'll bring it up here.

I'm a new game designer, you might be vaguely aware of my work if you've been to the dark corner of the internet called 4chan. I'm the lead designer of Song of Swords (we're launching soon) and Call of the Void. Around there I'm called Jimmy Rome.

Now I'm about 60 ounces into my Mickeys, otherwise I would never have had the courage to actually post here, so I'll get into it before I start sobering up.

I've got a question inspired by the Pundit's frequent rants on immersion and the damage that can be done to it by giving narrative control to players. And that question can only really be asked in the context of the game I'm making--Call of the Void.

One of the things we aim for in our games is making the combat feel real and the violence feel dangerous. To that end we have some of the most realistic and punishing damage tables out there, both for swords and such (SoS) and for boolets (CotV)

The problem is, bullets mess people up big-time. I'm a bit of a gun-nut myself, and I've shot a lot of rounds at a lot of things, and we really tried to represent the power of firearms of different calibers in-game.

The snag is, it's too easy for players to die relative to NPCs. Obviously X guy's meat isn't more bulletproof than Y guy's, that'd be dumb. My solution for this was to give players "Luck Points" where they would be able to cause attacks to miss them that would otherwise have hit them, either moving the attack to an NPC nearby, having the bullet wing an obstacle, or the gun jam, etc.

The question is--is this a design sin? Is giving the player an "out" when the dice turn against him a knife in the back of immersion? Will it crack that sense of tension when the player realize that the space nazi is aiming at him, to know that he can just spend a luck point and shrug it off?

Thanks for humoring me guys.

Quote from: Jimbojack;871159The question is--is this a design sin?

No it's not a design sin. It's a design choice. Like anything some people will enjoy having that in their game and some won't. Some sort of benny, like a luck point, is common in a lot of games.

QuoteIs giving the player an "out" when the dice turn against him a knife in the back of immersion?

It can be. The more the "out" or benny is something that corresponds to something in the game world that the character controls, the less unimmersive the "out" or benny will be. The more often a benny that isn't connected to the game world gets used and tracked during play the more unimmersive it will be.

In West End Games Star Wars D6 game there were Force Points (FP), a sort of luck point, that characters possessed. They allowed the character to double the number of dice rolled for that round. But since they were Force Points and we all had heard Ben Kenobi's voice saying "Use the Force" the use of an FP might still be considered immersive since it was something that was controlled by (or at least happening to) the character in game. So Luke uses the Force (spends 1 FP) to double his Star Ship Gunnery skill to send a proton torpedo down the Death Star's exhaust port. That tends to be pretty immersive for the player.

Contrast that with using a Fortune Point in Honor+Intrigue. A player might use a FP to add a bonus die to his attack roll. There isn't an analog to the Force in H+I so what the character is doing is less clear. We might say the PC is trying really hard to try to make spending 1 FP a little more connected to the character, but to an extent the FP represents the sort of lucky break we see in Heroes get in the swashbuckling genre which is, to an extent, character unimmersive, but often acceptable to a lot of players.

Contrast that further with a player in H+I whose character just fell, jumped, or was pushed out a window who spends 1 FP to have a convenient hay wagon underneath the window to break his fall. That is completely unrelated to anything the character does or controls. It is a narrative mechanic that allows the player some limited authorial control to allow the sort of lucky break that is scene in a lot of swashbuckling (and action) fiction. It isn't character immersive but it does fit a certain genre. Players will vary in how much they find the use of such a mechanic enhances the game and how much it detracts.

QuoteWill it crack that sense of tension when the player realize that the space nazi is aiming at him, to know that he can just spend a luck point and shrug it off?

It might. It depends on two things. First, how much the player's tension comes from the game mechanics and how much it comes from imagining the scene. So a player who strongly imagines his character staring down the barrel of a gun and that experience being scary might not think about the mechanical decrease in tension a luck point provides.

Second, typically a luck point is a scarce resource so shrugging off a shot from a space nazi isn't totally cost free. Rather than "cracking" or eliminating tension, a luck point is more likely to decrease the tension. The degree to which decreasing tension is a good or bad thing depends on on a lot of subjective player attitudes as well as how the mechanics actually work. It tends to be more acceptable to players in games that emulate genres like comic books, swashbuckling fiction, and Hollywood action movies than it is in games intended to be more life like and in which combat is supposed to be dangerous and rare rather than frequent and fun.

Quote from: Jimbojack;871159The question is--is this a design sin? Is giving the player an "out" when the dice turn against him a knife in the back of immersion?

Here's the thing.

If you check out how Joss works in Dangerous Journeys--most haters just sweep it aside as yet another run-of-the-mill deus ex machina mechanic like any old fate point or luck point system--you find it's actually an extremely "narrative" type of thing.

At first, I thought it would be a great and interesting addition to the game. It gave an example of the party captures, stripped of their gear, locked in cages, while the tribe of cannibals do some ceremonial dance in preparation for cooking the heroes. They want to use Joss to get out of the situation. The GM agrees, saying if everyone spends 2 Joss, they can escape.

The players spend the Joss, except two of the party only have 1 Joss left apiece. So the narrative effect is there's a sudden mudslide that washes over the tribe, as it tosses things about, the cages are broken open and the heroes are free... except for those two who only had 1 Joss, they're being swept away by the mud slide. It's now up to the rest of the party to save them.

Pretty cool, eh?

Except when I went to run the game, I found out it didn't work that way with my group. They would blow through all their Joss on dodge rolls and such to avoid capture in the first place. In other words, they used what you are regarding as a "narrative" mechanic as a simple extension of their characters' abilities.

I'm sure there might be some group out there that would use Joss for its awesome narrative influence. My point here is there ain't no such thing as a mechanic that's inherently narrative. If you give it as an option for players to use, they will use it in the manner of their choosing. If they don't care for narrative bullshit, they're not going to use it that way.


Incidentally, and maybe this idea could help you, in the Lejendary Adventure RPG, there is a Luck Ability. But it doesn't really work as luck points per se. It increases things like "saving throws" marginally. In combat, if you miss, you can opt for a "lucky hit" which just uses one-tenth of your Luck plus one-tenth of Weapons, with no other situational modifiers. We usually just opt not to because a) it slows down the game, and b) it introduces another chance for weapon breakage, so it's not worth it to use every single time, only when hitting is ultra important.

But probably the most common use of Luck is, once per round, you can attempt a "Lucky Dodge" of any attack. That's done with a simple Luck check. (Though if they are trying to dodge a critical hit, we assume you would need a critical success on your luck roll.) Something like that might suit your tastes better than having luck points. We've actually joked that if we wanted to simulate G.I. Joe, where bullets are flying everywhere but none of the main characters ever get hit, everyone has to have at least 90% Luck ability.

Some good commentary above and I am a bit pressed for time, but I'll add/emphasize:

--It's better for luck points to modify rolls or mitigate effects than it is for them to allow "definite" rolls or completely ignoring effects.

--It's also good to give them multiple uses so that they don't just function as a linear hit point buffer. E.g. allow their use "stumbling across clues".

--Yes, they do technically break "immersion" if by that you mean "in-character point of view". But they have some payoffs that might enhance immersion in some circumstances. For example, dying or being stuck in a dead-end may not be fun, and could cause players to become detached from the game.

--Methods of gaining points are as important as methods of using them. It's also worth considering if players should retain points from dead/retired characters, to use on their new character.

Quote from: Jimbojack;871159Now I'm about 60 ounces into my Mickeys, otherwise I would never have had the courage to actually post here, so I'll get into it before I start sobering up.

Post away whenever you like - People can be friendly here, at times.

Quote from: Jimbojack;871159The snag is, it's too easy for players to die relative to NPCs. Obviously X guy's meat isn't more bulletproof than Y guy's, that'd be dumb. My solution for this was to give players "Luck Points" where they would be able to cause attacks to miss them that would otherwise have hit them, either moving the attack to an NPC nearby, having the bullet wing an obstacle, or the gun jam, etc.

The question is--is this a design sin? Is giving the player an "out" when the dice turn against him a knife in the back of immersion? Will it crack that sense of tension when the player realize that the space nazi is aiming at him, to know that he can just spend a luck point and shrug it off?

In my opinion, as soon as you roll dice or look at a table then you have lost immersion. When I play, I tend to suspend immersion, do something gamesy then unsuspend immersion and go with whatever the gamesy thing did.

So, Luck Points are fine with me. I see them as Arnie being able to escape with just a flesh wound even though he has had an uzi firing at him.

The tension isn't when you have lots of Luck Points, but when they are running pout very quickly. If you are running low then every combat can be dangerous, or can seem dangerous.

It means that Luck Points are a managed resource. Do you save them for later and risk incurring a wound, or do you avoid this wound and then fall victim to a more serious wound later on?

I'd say keep them in as you seem to have a good idea.

The whole point of SoS/CotV-nastiness is to not get hit, yes? If players had a limited pool of fate/luck/hero/etc points, they could use them to dive behind cover or perhaps carry a convinient copy of Hero6e. Alternatively, that bullet in the gut happened to pass through it all without actually hitting anything really bad. It will still hurt like hell, but you'll live - at least for now.
I see nothing wrong or off-putting with having a mechanic for "getting out of jail". It all depends on how you incorporate it into the game - is it a finite resource (like Warhammer's fate points), something that regenerates between sessions or in-game time (like RQ's luck points) or a continous and ill-conceived economy of fate points (like 2d20)? And what should they do? Allow for PC's to avoid certain death? Allow PC's a second chance to avoid certain death, horrendous injury or unheroic defeat? Ensure insta-win?

This is a thorny subject, but really worth debating...

Quote from: Arminius;871202--It's better for luck points to modify rolls or mitigate effects than it is for them to allow "definite" rolls or completely ignoring effects.

I have the opposite opinion, unless there is a very neat way in your game to "mitigate" effects of wounds. The most common use of Luck Points is to allow a reroll: however, this can be almost as baffling as leaving the thing completely to luck. I have seen a player choose a re-roll over a "bump" of level of success because he had rolled a 99 on percentile, he did not expect to reroll a fumble. He rolled 99 again. I have seen a character die after spending his entire allotment of Luck Points for the session because he missed the necessary rolls four times in a row.

The point of Luck Points is to avoid realistic-but-anti-climactic events. Re-rolls and half-baked solutions do not guarantee that you avoid them: on the other hand, when they fail they stress the fact that the event was anti-climactic, as the player becomes more emotionally invested in the outcome after spending a limited resource on it.

So, instead of nerfing the effect of luck, just stress the fact that the character will not be that lucky next time by limiting the available luck.

Quote--Methods of gaining points are as important as methods of using them.

THIS is never stressed enough. My model for how things should work is Marvel Superheroes (FASERIP), a 1983 TSR game - is it old school enough?. Not only does it force you to decide which rolls are so significant for your hero to waste oh-so-precious Karma points on them, but it links Karma to behaviour in a very effective way, letting you gain it only by being a hero, in a silver/bronze age way: doing charity for children, sparing lives, being generous etc. No "earn luck because you introduced some entertaining details in the game" bullshit: you are there to play a hero, not an entertainer. And there is no "luck that regenerates between sessions", what you have is what you earned by being the hero that society expects you to be. And if you kill someone, even by mistake, you lose all Karma: not 100 points, not 1000 - ALL. If you wanted to play the Punisher or Wolvie with that game, all of the killing would take place off-screen. Like it happened in the 1980 comic books.

--------------

In general, Luck points are a solution that most players will find acceptable, in one way or another. Some will accept them only when there is an "in-game" justification for them (The Force in Star Wars, as suggested), others will not mind. But people who really cannot stand them in any form are rare - they exist, but they are few.

The great thing about Luck Points in game is that they work on a different plane than physics. To ensure that revolvers will not kill PCs outright with a bad roll, you have basically two solutions: either you scramble the world physics so that every gun is as effective as a .22 used with Stormtrooper-level accuracy, or you bend probability in favour of the heroes. Between altered bullet effects on bodies or altered probabilities, I think there is not much debate on what has the most detrimental effect on suspension of disbelief.

----------------

As a final note: someone could think that Luck points are a storygame thing. Usually, the ones who are most vocal in their crusade against the Forge (the Pundit, or Brent Dedeaux from rpg.net's Tales from the Rocket House) also hate Fate Points with all their heart.

The point is that Luck Points have nothing to do with forgey games. Forgies consider them a rudimentary mechanics for addressing theme and creating "story". All those supposedly-forgie games that use them have been designed before the Big Model came out, and have been attached to the Forge theory at a second time. Not surprisingly, they are the Forge games that most resemble classic RPGs (Riddle of Steel, Burning Wheel, etc.). Luck and Fate are more of a "fix" applied to classic or even OS games where you want to limit lethality and anti-climactic results, rather than a storygame thing.

My advice is to just make the game you want to make. However you might want to do a search on genre emulation on this site as Pundit has written a number of posts on that subject from a non-immersion breaking point of view.

If you don't want to break immersion, and you agree with a lot of the posters here that one of the things that breaks immersion is 'stepping out of your character's headspace', consider making anything like Luck points that might protect player characters from harm an aspect of the setting itself. That is still going to be  potential issue for people but if there is an in game reason for the luck points and they represent something that the player character can handle without stepping being a floating resource pool the character is unaware of, that might help (i.e. if people can literally call upon a patron saint or something a few times a day for a little help....it is less like I am spending abstract points and takes it into being something real my character and I both conceptualize in the same way). I'd probably not even make it luck points, but just have it be a system where it gets progressively harder to invoke the bonus because you are annoying the saint in question (or whatever it is that is bestowing the aid). Things like that can help. Also a zero to hero system can help as well. If the power levels between low level characters and more veteran characters is wide enough that really helps survivability if that is a concern.

Quote from: RosenMcStern;871344I have the opposite opinion, unless there is a very neat way in your game to "mitigate" effects of wounds.

James Bond 007 would be the classic example of mitgating (or increasing) wound levels via Hero Points. Also one of the earliest examples in a published game (1983).

QuoteI have seen a player choose a re-roll over a "bump" of level of success because he had rolled a 99 on percentile, he did not expect to reroll a fumble. He rolled 99 again.

This isn't exclusive to systems that use roll over. It's a function of, as you point out, uncertainty or randomness in the effect of the benny. Players are unhappy when they spend their benny point and the net effect is negligible, nothing, or even a worse result on the re-roll than the first roll. I think this frustration is akin to players who get frustrated in systems with separate rolls for attack and for damage. If a player rolls a great attack roll (maximal or near maximal) and then rolls a lame (e.g. minimal or near minimal) damage roll, frustration frequently ensues. It can be even worse in a system that uses armor to absorb or block damage as a great attack roll may result in no damage penetrating armor. Which many players find frustrating. And yet, systems that use more than one roll to determine combat results are the systems that are most commonly played.

QuoteTHIS is never stressed enough. My model for how things should work is Marvel Superheroes (FASERIP), a 1983 TSR game - is it old school enough?. Not only does it force you to decide which rolls are so significant for your hero to waste oh-so-precious Karma points on them, but it links Karma to behaviour in a very effective way, letting you gain it only by being a hero, in a silver/bronze age way: doing charity for children, sparing lives, being generous etc. No "earn luck because you introduced some entertaining details in the game" bullshit: you are there to play a hero, not an entertainer. And there is no "luck that regenerates between sessions", what you have is what you earned by being the hero that society expects you to be. And if you kill someone, even by mistake, you lose all Karma: not 100 points, not 1000 - ALL. If you wanted to play the Punisher or Wolvie with that game, all of the killing would take place off-screen. Like it happened in the 1980 comic books.

Which is fine if we are playing 80s superheroes. (I game I have never, ever played.) It ignores the types of characters that the vast majority of other games allow who are not nice-guy, 1980s comic book superheros. That's probably one reason why Gerard Christopher Klug didn't tie gaining a Hero Point to being a nice guy in James Bond 007. Mr. Bond is really not a nice guy.

If the GM doesn't want to incentivize one behavior (like nice guy 1980s comics morality) over another than some other method of generating Hero, Luck, Fortune, or whatever we choose to call those points is required.

As a side note, your disdain for subjective "earn luck because you introduced some entertaining details in the game" seems a little odd given that you laud the equally subjective "play a hero." Both are subjective measures and either can be used to reinforce a certain genre or tone at the table. Preferring 80s comic books over any other genre is similarly a subjective preference. Handing out Luck points for wise-cracking dialog or Bond-like bon mots reinforces a genre in much the same way that handing out Karma for helping a little old lady across the street or delaying your chase of Doc Ock to get a lost kitten out of a tree.

QuoteBetween altered bullet effects on bodies or altered probabilities, I think there is not much debate on what has the most detrimental effect on suspension of disbelief.

I think you are wrong. Either can be equally suspension breaking. Star Wars Stormtroopers who can't hit the heroes who are standing still in the open, action film heroes who routinely outrun traversing machineguns and strafing planes or have fusillades of bullets all miraculously swerve around them, and WWII German soldiers who routinely pause in their run from cover to cover so that Sgt. Saunders and his men can get a clean shot are no less immersion breaking than is John McClain shrugging off yet another fall, beating, laceration, or bullet wound.

QuoteAs a final note: someone could think that Luck points are a storygame thing.

Using a benny to alter the game world by having a hay wagon underneath the window to break the heroes fall or to automatically get a clue (whether or not there was one there in the first place) is an authorial tool giving the player one time control over the environment. That has been a use of bennies since at least 1983. I don't care to ride Pundy's hobby horse, but I can see why someone might characterize such a device as being more story game than not.

QuoteLuck and Fate are more of a "fix" applied to classic or even OS games where you want to limit lethality and anti-climactic results, rather than a storygame thing.

If you limited the use of bennies to limiting lethality only, I'd be more likley to agree. But avoiding anticlimax opens up the usage of bennies outcomes that can be story oriented. A result that makes for a bad "story" might be considered anti-climactic. Using bennies to avoid anti-climax is a way of allowing the player to shape the "story" in a way that they think they will find more satisfying. Which seems kind of story gamey to me.

Not a sin, just a choice. It's not my cup of tea, but it can work. It sounds like a good workaround for gun damage. If it works, it works.

//Panjumanju

Quote from: Bren;871355I think this frustration is akin to players who get frustrated in systems with separate rolls for attack and for damage. If a player rolls a great attack roll (maximal or near maximal) and then rolls a lame (e.g. minimal or near minimal) damage roll, frustration frequently ensues.

I doubt it is the same thing. You can re-roll the attack next turn. The bennie however, is gone and wasted forever (or till next session at most). I had clearly specified that "the player becomes more emotionally invested in the outcome after spending a limited resource on it."

QuoteAs a side note, your disdain for subjective "earn luck because you introduced some entertaining details in the game" seems a little odd given that you laud the equally subjective "play a hero." Both are subjective measures and either can be used to reinforce a certain genre or tone at the table. Preferring 80s comic books over any other genre is similarly a subjective preference.

I never stated that I consider the 80s comic books as the One True Way to be a hero. I stressed that FASERIP clearly defined what a hero is in that context, by using a well-known behavioural model and eliminating subjectivity. Naturally, if you want to play another kind of cool character that ruleset should be at least hacked. Playing Watchmen with it wouldn't be possible, for instance. Yet it would be cool to play a Watchmen game.

QuoteHanding out Luck points for wise-cracking dialog or Bond-like bon mots reinforces a genre in much the same way that handing out Karma for helping a little old lady across the street or delaying your chase of Doc Ock to get a lost kitten out of a tree.

Similarly, I never wrote that assigning bennies for "genre appropriate" action is wrong unless it follows the esthetic criteria of Chris Claremont or Roy Thomas. If you read my own games, they include plenty of Fate granted for way less "heroic" actions! The point is that the game should clearly define what is genre/setting appropriate and rewarded with "fuel" for your heroic actions. Just stating "when the GM thinks it is fun" is an incomplete work on the part of the designer.

Basically, I was trying to keep self-promotion out of the door and use a well-known old-school ruleset instead of quoting my own work Please do not take it as a "FASERIP is the only true way to go" statement. The point is that a well-designed luck point economy needs to be heavily tailored for the genre you are playing.

Quote from: RosenMcStern;871372I doubt it is the same thing. You can re-roll the attack next turn. The bennie however, is gone and wasted forever (or till next session at most).

Its the same in the sense that both frustrations come out of an expectation that something (expending a Hero Point or rolling really well with the dice) will yield an anticipated better result than what actually occurs in play. I'm not trying to say that expending your very last Hero Point only to get nothing results in the same intensity of frustration as rolling really well in combat only to have the effective damage be minimal or zero, only that it is the same kind of frustration. Watch a player who has repeatedly missed in combat finally roll a successful hit - and a really good to hit roll at that - only to have the damage roll give 1 point of damage or none and you will often see considerable, and I would say similar, frustration. Now you might say, well they can hope to roll better next round (if there is a next round). But by the same token, virtually every system with a hero/luck point mechanic has some means of refreshing points or gaining new points.

QuoteI had clearly specified that "the player becomes more emotionally invested in the outcome after spending a limited resource on it."

And I am saying that often the player who rolls a really good attack roll is more emotionally invested in the effective damage that results being above the minimum than is a player who succeeds, but whose success roll just barely succeeds. The cause is the frustration of expectations in a situation where outcome matters to the player. That applies both to spending Hero points and getting little or nothing better as it does to good attack rolls that have little actual effect.

QuoteI never stated that I consider the 80s comic books as the One True Way to be a hero. I stressed that FASERIP clearly defined what a hero is in that context, by using a well-known behavioural model and eliminating subjectivity.

I seriously doubt it eliminates all subjectivity. As I recall* it provides guidelines that make the awards less subjective, but I believe the GM still has some room for subjective judgement. So do many (if not most) games that include bennies, including, but not limited to James Bond 007, WEG Star Wars, and Honor+Intrigue to name just three.

QuoteNaturally, if you want to play another kind of cool character that ruleset should be at least hacked. Playing Watchmen with it wouldn't be possible, for instance. Yet it would be cool to play a Watchmen game.

At the point you have to hack it, it isn't generally applicable. Which was part of what I was pointing out. Clearly you agree. I'm not seeing that FASERIP provides a better mechanic than lots of other systems, though I accept that you like it better.

QuoteSimilarly, I never wrote that assigning bennies for "genre appropriate" action is wrong unless it follows the esthetic criteria of Chris Claremont or Roy Thomas.

Great then we are in agreement about that.

QuoteThe point is that the game should clearly define what is genre/setting appropriate and rewarded with "fuel" for your heroic actions. Just stating "when the GM thinks it is fun" is an incomplete work on the part of the designer.

On that we do not agree. I expect the designer to provide guidelines. I'm perfectly capable of creating my own or changing theirs, but being somewhat lazy about creating game mechanics I prefer the rules include guidelines. Of course the three games I mentioned all do that.

QuoteBasically, I was trying to keep self-promotion out of the door and use a well-known old-school ruleset instead of quoting my own work Please do not take it as a "FASERIP is the only true way to go" statement. The point is that a well-designed luck point economy needs to be heavily tailored for the genre you are playing.

Your lack of self promotion is refreshing.

I agree that the benny economy needs to be tailored, though I'd say it is far more important that it be tailored to the table preferences than to the game designer's notion of genre. So in theory, I want more, not less subjectivity in the how the economy is managed so that it can be managed to the preferences of any given table. In practice, I want guidelines from the designer that are informed by multiple play tests with different groups and GMs.


* I could be wrong, it's been a long time since I looked at the rules which are probably somewhere in my basement. I note in passing that James Bond includes an almost entirely objective measure "Characters earn a Hero Point every time they get a Quality 1 result on a skill other than combat." But even that is subjective in that the GM decides when a roll is required to be made. And the game does include an entirely subjective award "when the GM chooses to award one for a clever or dramatic action."

YMMV but I feel that in a comparison between several uses of Luck:

--Declare beforehand for a bonus on the dice (or perhaps for "roll twice, take better result")

--Declare afterward for reroll (or perhaps "reroll and use best result, but no better than a normal success")

--Declare beforehand and get the exact roll you want (including critical success)

The last to me seems anticlimactic. Now I'm probably stacking the deck a bit with my examples but the point is that some level of risk is often (usually?) fun. So you have to balance the benefit of luck points in terms of smoothing out unfortunate anticlimactic effects or enabling the players to "push" for exciting climactic successes, against the fact that they can also take the excitement out of a climactic action by making it a sure thing.

I'm sure there are many approaches available to fine tune "luck" to taste. One of the great things about its use in trad games is that you can usually ignore it or modify it, giving you everything from brutal skirmish wargame to Hollywood cinematics.

I'm generally not fond of 'Luck Points'... my first experiences of them was in a game where they were so plentiful and versatile that they pretty much removed any chance of failure. Probably the GM's fault but whatever...

I'm also generally not fond of trying to game the sort of source material they're usually aimed at emulating... movie action heroes dodging bullets and explosions and having plot immunity.

I don't mind them so much if they're somehow accounted for in-setting and don't guarantee success... and aren't so plentiful or easily regained that they're used even for casual stuff.
The way they work in DCC doesn't bother me much because they're pretty much an representation of actual 'luck' in the game as a measurable quantity, like magic. They're limited by the GM's agency, and using them can have serious repercussions later when you're caught with your Luck down.
I can also see them in something like Runequest as a representation of the favor/disfavor of your chosen diety.

The idea of spending the BEFORE a roll appeals to me as well... like, "I'm gonna bet the farm on this attempt... gonna try really hard and go for it." rather than some sort of (lame IMO) do-over.

I certainly do not want them in horror games like Call of Cthulhu or Kult.

I think you have to be really careful with a form of luck that "automatically works."

Suppose you have some plan so crazy it just might work. But it boils down to two key skill rolls where the dice just might go south on you. If a luck point can give you an automatic success and you have two luck points, you don't have to sweat this. You don't have to bother with contingencies. Luck used in this way can be "destructive" in the sense it alters behavior and decision-making.

Luck that gives a re-roll, though, still not guaranteeing success? That's fine. The plan still requires a little luck, and you can still use luck to substantially increase the odds of success. But you STILL need to plan for contingencies. In this way, luck does function to boost the awesome of player characters without altering fundamental decision-making. That's how you can achieve your goals of giving PCs a much better shot at surviving without altering immersion. (I'm measuring "immersion" by the only thing that matters--decision-making. I don't count getting distracted. A million things from real life could distract you while you're gaming. A reasonable departure, like to reference a rule, does not phase a good, immersive player.)